Online edits and Scan time [SLC 5/05]

Can adding in online edits affect the scan time of a SLC 5/05?

I'm re-writing a section of logic in one right now, and before I started, the scan time was on average around 8 msecs. I added in my 21 rungs, and accepted them, and marked 36 for deletion. I haven't even put edits into test mode and checking the scan time it is averaging somewhere around 23 msecs.

I'm just concerned because I'm using S:4/0 (low bit of the free running clock with a 20 msec cycle) as a psuedo-pulse train for a tracking algorithm and need my scan time to be below 20 msecs for it to work.

It just seems odd to me that doing nothing but adding in some accepted but untested rungs is putting such a toll on my scan time.

Any ideas?

When you accept the rungs they are downloaded into the PLC but not yet scanned. I cannot imagine that there is not some significant overhead involved with selecting which version of the rung is to be scanned and which is not, especially since you currently have 57 rungs under edit. Until you assemble those edits the memory occupied by the rungs (both new and old copy) in the PLC is not contiguous, despite how it appears in your editor - you are doing quite a bit of hopping around.

Thanks Alaric. I knew the rungs were downloaded to the PLC once I accepted them, but I didn't think about the memory being non-contiguous.

When the production folks clear out, I'm going to assemble the edits and see if that squares it away.

Yep... after assembling, the scan time dropped right back down to where it was before said edits started.

Now I just need to figure out why the hell my tracking algorithm isn't staying consistent. Maybe I'll try a timed interrupt instead of shifting my tracking index on the rising edge of S:4/0.